What LLMs Fit in 16 GB of VRAM? (Oct 2026)

Radeon RX 7800 XT 16GB (desktop) is the catalog SKU in this 16 GB tier with the smallest net budget: 14.00 GiB after the sourced reserve. 44 catalog models fit at a file-backed artifact. The month follows dataset 2026-10-02.1 (Oct 2026), not the build clock.

Catalog dataset 2026-10-02.1 (Oct 2026).

Catalog SKUs in this tier

SKUMarketed capacityNet budget
Radeon RX 7800 XT 16GB (desktop)
Smallest net budget
16 GB GDDR614.00 GiB
Radeon RX 9070 16GB (desktop)16 GB GDDR614.00 GiB
Radeon RX 9070 XT 16GB (desktop)16 GB GDDR614.00 GiB
GeForce RTX 4080 16GB (desktop)16 GB GDDR6X14.00 GiB
GeForce RTX 4080 SUPER 16GB (desktop)16 GB GDDR6X14.00 GiB
GeForce RTX 5060 Ti 16GB (desktop)16 GB GDDR714.00 GiB
GeForce RTX 5070 Ti 16GB (desktop)16 GB GDDR714.00 GiB
GeForce RTX 5080 16GB (desktop)16 GB GDDR714.00 GiB
GeForce RTX 4060 Ti 16GB (desktop)16 GB GDDR615.00 GiB
GeForce RTX 4070 Ti SUPER 16GB (desktop)16 GB GDDR6X15.00 GiB

Catalog models that fit

Each row is one file-backed artifact on the representative SKU. A missing quantization is omitted, not replaced. Context search stops at the sourced cap. With no sourced cap, the only verdict shown is at 8192 tokens.

When catalog bandwidth is missing, the decode ceiling says not available. That is not a memory verdict.

ModelArtifactMax context that fitsMax context including tightVerdict at 8192 tokensDecode ceilingLinks
Qwen2.5-0.5B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
0.68 GiB–1.07 GiB
Not available
Qwen3-0.6BGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
40,96040,960
8192
Runs comfortably
1.37 GiB–1.81 GiB
Not available
Gemma 3 1B ITGGUF Q4_0 (Google QAT)
gguf-q4_0 · llama.cpp
32,76832,768
8192
Runs comfortably
1.10 GiB–1.50 GiB
Not available
Llama 3.2 1B InstructOfficial BF16 safetensors
bf16 · vllm
131,072131,072
8192
Runs comfortably
3.18 GiB–5.06 GiB
Not available
Qwen2.5-1.5B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
1.38 GiB–1.80 GiB
Not available
Qwen2.5-Coder-1.5B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
1.38 GiB–1.80 GiB
Not available
SmolLM2-1.7B-InstructGGUF Q4_K_M (Hugging Face)
gguf-q4_k_m · llama.cpp
8,1928,192
8192
Runs comfortably
2.61 GiB–3.11 GiB
Not available
Qwen3-1.7B-BaseGGUF Q4_K_M (mradermacher)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
2.03 GiB–2.49 GiB
Not available
Qwen3-1.7BGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
40,96040,960
8192
Runs comfortably
2.03 GiB–2.49 GiB
Not available
Qwen2.5-3B-InstructGGUF Q4_K_M
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
2.37 GiB–2.80 GiB
Not available
Qwen2.5-Coder-3B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
2.37 GiB–2.80 GiB
Not available
Llama 3.2 3B InstructOfficial BF16 safetensors
bf16 · vllm
37,41069,194
8192
Runs comfortably
7.48 GiB–9.39 GiB
Not available
Qwen2.5-VL-3B-InstructGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
128,000128,000
8192
Runs comfortably
2.20 GiB–2.63 GiB
Not available
Phi-3.5-mini-instructGGUF Q4_K_M (bartowski)
gguf-q4_k_m · llama.cpp
25,07731,801
8192
Runs comfortably
5.35 GiB–5.98 GiB
Not available
Phi-4-mini-instructGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
74,51594,652
8192
Runs comfortably
3.45 GiB–3.94 GiB
Not available
Qwen3-4BGGUF Q4_K_M
gguf-q4_k_m · llama.cpp
40,96040,960
8192
Runs comfortably
3.58 GiB–4.08 GiB
Not available
Qwen3-4B-Instruct-2507GGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
66,20084,098
8192
Runs comfortably
3.58 GiB–4.08 GiB
Not available
Gemma 3 4B ITGGUF Q4_0 (Google QAT)
gguf-q4_0 · llama.cpp
131,072131,072
8192
Runs comfortably
3.33 GiB–3.79 GiB
Not available
Qwen3-VL-4B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
66,20084,098
8192
Runs comfortably
3.58 GiB–4.08 GiB
Not available
Mistral-7B-Instruct-v0.3GGUF Q4_K_M (bartowski)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
5.20 GiB–5.72 GiB
Not available
DeepSeek-R1-Distill-Qwen-7BGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
131,072131,072
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
OpenMath-Nemotron-7BGGUF Q4_K_M (mradermacher)
gguf-q4_k_m · llama.cpp
131,072131,072
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
Qwen2-7B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
Qwen2.5-7B-InstructGGUF Q4_K_M
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
Qwen2.5-Coder-7BGGUF Q4_K_M (mradermacher)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
Qwen2.5-Coder-7B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
Qwen2.5-Math-7B-InstructGGUF Q4_K_M (bartowski)
gguf-q4_k_m · llama.cpp
4,0964,096
8192
Runs comfortably · Rough estimate
4.92 GiB–6.23 GiB
Not available
DeepSeek-R1-Distill-Llama-8BGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
56,85076,121
8192
Runs comfortably
5.71 GiB–6.24 GiB
Not available
granite-3.3-8b-instructGGUF Q4_K_M (IBM)
gguf-q4_k_m · llama.cpp
45,35160,762
8192
Runs comfortably
5.98 GiB–6.53 GiB
Not available
Qwen3-8BGGUF Q4_K_M
gguf-q4_k_m · llama.cpp
40,96040,960
8192
Runs comfortably
5.93 GiB–6.48 GiB
Not available
Qwen3-8B-BaseGGUF Q4_K_M (mradermacher)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
5.93 GiB–6.48 GiB
Not available
Qwen2.5-VL-7B-InstructGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
128,000128,000
8192
Runs comfortably
4.92 GiB–5.42 GiB
Not available
Qwen3-VL-8B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
49,84166,937
8192
Runs comfortably
5.93 GiB–6.48 GiB
Not available
GLM-4-9B-0414GGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
32,76832,768
8192
Runs comfortably
6.18 GiB–6.69 GiB
Not available
Gemma 3 12B ITGGUF Q4_0 (Google QAT)
gguf-q4_0 · llama.cpp
62,70799,001
8192
Runs comfortably
8.46 GiB–9.04 GiB
Not available
Mistral-Nemo-Instruct-2407GGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
30,60645,293
8192
Runs comfortably
8.34 GiB–8.94 GiB
Not available
phi-4GGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
16,38416,384
8192
Fits
9.97 GiB–10.62 GiB
Not available
Phi-4-reasoningGGUF Q4_K_M (Unsloth)
gguf-q4_k_m · llama.cpp
17,14928,540
8192
Fits
10.12 GiB–10.77 GiB
Not available
Qwen3-14BGGUF Q4_K_M
gguf-q4_k_m · llama.cpp
21,73535,988
8192
Fits
9.76 GiB–10.39 GiB
Not available
OpenMath-Nemotron-14BGGUF Q4_K_M (mradermacher)
gguf-q4_k_m · llama.cpp
18,17930,060
8192
Fits
10.00 GiB–10.64 GiB
Not available
Qwen2.5-14B-InstructGGUF Q4_K_M
gguf-q4_k_m · llama.cpp
18,17930,060
8192
Fits
10.00 GiB–10.64 GiB
Not available
Qwen2.5-Coder-14B-InstructGGUF Q4_K_M (Qwen)
gguf-q4_k_m · llama.cpp
18,17930,060
8192
Fits
10.00 GiB–10.64 GiB
Not available
DeepSeek-V2-Lite-ChatGGUF Q4_K_M (mradermacher)
gguf-q4_k_m · llama.cpp
17,137145,761
8192
Fits · Rough estimate
10.01 GiB–12.23 GiB
Not available
Qwen3.8-27BGGUF UD-IQ1_S (Unsloth)
gguf-ud-iq1_s · llama.cpp
67,617130,418
8192
Runs comfortably · Rough estimate
6.54 GiB–8.14 GiB
Not available

Allowlisted combos